Losing somatic SNPs

Question

[gab]

I got circa 130 high confidence somatic SNP, comparing tumor and blood samples with VarScan. I wanted to annotate them so I passed the data to VarAft, but it gives me just 6 variations. What am I doing wrong? Is there a tool that gives me the info i can get with VarAft?

[Denise CS]

Have you thought in annotating this file with other tool than VarAft. If it works for all your variants, the problem is with VarAft. It it works for your 6 variants only, the problem is in your input file.

It may be helpful to give us a sample of your somatic variants here.

[gab]

chrom  position   ref  var  normal_reads1  normal_reads2  normal_var_freq  normal_gt  tumor_reads1  tumor_reads2  tumor_var_freq  tumor_gt  somatic_status  variant_p_value  somatic_p_value        tumor_reads1_plus  tumor_reads1_minus  tumor_reads2_plus  tumor_reads2_minus  normal_reads1_plus  normal_reads1_minus  normal_reads2_plus  normal_reads2_minus
1      1268181    T    C    36             0              0%               T          9             2             18,18%          Y         Somatic         1.0              0.050878815911191914   9                  0                   2                  0                   35                  1                    0                   0
1      17084510   G    A    148            7              4,52%            G          67            13            16,25%          R         Somatic         1.0              0.0031524490371238824  67                 0                   13                 0                   140                 8                    7                   0
1      26887584   C    A    19             0              0%               C          8             2             20%             M         Somatic         1.0              0.11083743842364542    8                  0                   2                  0                   19                  0                    0                   0
1      27873916   G    T    45             0              0%               G          9             2             18,18%          K         Somatic         1.0              0.035714285714286274   9                  0                   2                  0                   41                  4                    0                   0
1      45504720   C    G    52             3              5,45%            C          35            14            28,57%          S         Somatic         1.0              0.001473746031299      34                 1                   14                 0                   50                  2                    3                   0
1      57221553   T    G    12             0              0%               T          3             9             75%             G         Somatic         1.0              1.682595234890286E-4   3                  0                   9                  0                   11                  1                    0                   0
1      57221554   G    A    13             0              0%               G          3             9             75%             A         Somatic         1.0              1.076860950329785E-4   3                  0                   9                  0                   12                  1                    0                   0
1      92467622   C    T    8              0              0%               C          13            2             13,33%          Y         Somatic         1.0              0.4150197628458471     13                 0                   2                  0                   8                   0                    0                   0

This is some example of .snp output from VarScan. Annotating with VEP gives no problems.

[Denise CS]

Are these coordinates on GRCh37 or GRCh38? If the VEP runs fine with your input file, the chances are there is something going on with VarScan.

Can you compare the 6 positions that work with VarScan with the ones that do not work? What is different between this two groups?

[gab]

It turned out bam files were generated with a different fasta instead of the one I used to get the mpileups, this lead to indexing problems and all SNPs were misplaced. Tahnks to everybody for the help!

[ATpoint]

Does the tool remove variants that are unlikely to be related to disease? If so, I would not be surprised that the majority of mutations got eliminated. Most somatic mutations have no causality towards the disease but are passengers.